Cheng Peng

In this visualization our goal is to find the factor that has the strongest correlation with salary. So I first ran a simple linear regression model on the three factors, then picked the r^2 to visualize because it tells us how good the fit is. From the small r^2 value we could tell that the LINEAR correlation is weak for all three factors. Although using a polynomial instead of simple fit could result a much better fit, the issue gets more complicated since the degree of the polynomial would vary from factor to factor.

* To see the detail it's easier to click on the image and zoom in.

I laid out the distributions horizontally so it's possible to compare the same age group among different work types.

* To see the detail it's easier to click on the image and zoom in.

1. Largest overall sales can be seen from the bar that represents the monthly average revenue;

2. The regions with growing sales revenue can be seen immediately from the regression line and color of bar;

3. The region that trends up best is obvious from both the regression line and the saturation of colors;

4. Comparison between different regions can be made easily since they are laid out horizontally.